English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Source-free domain adaptation (SFDA) aims to adapt models trained on alabeled source domain to an unlabeled target domain without the access tosource data. In medical imaging scenarios, the practical significance of SFDAmethods has been emphasized due to privacy concerns. Recent State-of-the-artSFDA methods primarily rely on self-training based on pseudo-labels (PLs).Unfortunately, PLs suffer from accuracy deterioration caused by domain shift,and thus limit the effectiveness of the adaptation process. To address thisissue, we propose a Chebyshev confidence guided SFDA framework to accuratelyassess the reliability of PLs and generate self-improving PLs forself-training. The Chebyshev confidence is estimated by calculating probabilitylower bound of the PL confidence, given the prediction and the correspondinguncertainty. Leveraging the Chebyshev confidence, we introduce twoconfidence-guided denoising methods: direct denoising and prototypicaldenoising. Additionally, we propose a novel teacher-student joint trainingscheme (TJTS) that incorporates a confidence weighting module to improve PLsiteratively. The TJTS, in collaboration with the denoising methods, effectivelyprevents the propagation of noise and enhances the accuracy of PLs. Extensiveexperiments in diverse domain scenarios validate the effectiveness of ourproposed framework and establish its superiority over state-of-the-art SFDAmethods. Our paper contributes to the field of SFDA by providing a novelapproach for precisely estimating the reliability of pseudo-labels and aframework for obtaining high-quality PLs, resulting in improved adaptationperformance.

A Chebyshev Confidence Guided Source-Free Domain Adaptation Framework for Medical Image Segmentation